A Superpixel-Based Algorithm for Detecting Optical Density Changes in Choroidal Optical Coherence Tomography Images...

Area of Science:

• Ophthalmology
• Medical Imaging
• Biomarkers

Background:

• Systemic diseases can manifest with vascular changes detectable in ocular tissues.
• Optical coherence tomography (OCT) is a non-invasive imaging technique for visualizing ocular structures.

Purpose of the Study:

• To investigate the diagnostic potential of image-processing analysis in OCT images.
• To detect systemic vascular changes in individuals with systemic diseases, specifically diabetes.

Main Methods:

• Analysis of ocular OCT images from diabetic patients and healthy controls.
• Application of a novel Superpixel Segmentation (SpS) algorithm to extract optical image density from ocular vascular tissue.
• Assessment of choroidal area (CA) and choroidal optical image density (COID) as metrics for vascular changes.

Main Results:

• Significant differences in CA and COID were observed between diabetic and healthy eyes.
• These parameters (CA and COID) show potential as valuable biomarkers for early vascular damage.
• The SpS algorithm demonstrated good inter- and intra-observer repeatability.

Conclusions:

• Digital image-processing techniques applied to OCT B-scan images can identify new parameters linked to ocular vascular damage.
• These findings suggest novel indicators of pathology for systemic diseases.
• OCT image analysis offers a promising approach for early detection of vascular alterations.